View
3
Download
0
Category
Preview:
Citation preview
Active MEMS metamaterials for THz bandwidth controlKailing Shih, Prakash Pitchappa, Manukumara Manjappa, Chong Pei Ho, Ranjan Singh, Bin Yang, Navab Singh,and Chengkuo Lee
Citation: Appl. Phys. Lett. 110, 161108 (2017); doi: 10.1063/1.4980115View online: http://dx.doi.org/10.1063/1.4980115View Table of Contents: http://aip.scitation.org/toc/apl/110/16Published by the American Institute of Physics
Articles you may be interested in Controlling phase of microwaves with active graphene surfacesAppl. Phys. Lett. 110, 161102161102 (2017); 10.1063/1.4980087
Sensing with toroidal metamaterialAppl. Phys. Lett. 110, 121108121108 (2017); 10.1063/1.4978672
Silicon-immersed terahertz plasmonic structuresAppl. Phys. Lett. 110, 151105151105 (2017); 10.1063/1.4980018
Dielectric metasurfaces for quantum weak measurementsAppl. Phys. Lett. 110, 161115161115 (2017); 10.1063/1.4982164
Designing an efficient rectifying cut-wire metasurface for electromagnetic energy harvestingAppl. Phys. Lett. 110, 083901083901 (2017); 10.1063/1.4976804
Broadband metamaterial for optical transparency and microwave absorptionAppl. Phys. Lett. 110, 143511143511 (2017); 10.1063/1.4979543
http://oasc12039.247realmedia.com/RealMedia/ads/click_lx.ads/www.aip.org/pt/adcenter/pdfcover_test/L-37/549665608/x01/AIP-PT/APL_ArticleDL_042617/FYI_HouseBanner_1640x440_v3.jpg/434f71374e315a556e61414141774c75?xhttp://aip.scitation.org/author/Shih%2C+Kailinghttp://aip.scitation.org/author/Pitchappa%2C+Prakashhttp://aip.scitation.org/author/Manjappa%2C+Manukumarahttp://aip.scitation.org/author/Ho%2C+Chong+Peihttp://aip.scitation.org/author/Singh%2C+Ranjanhttp://aip.scitation.org/author/Yang%2C+Binhttp://aip.scitation.org/author/Singh%2C+Navabhttp://aip.scitation.org/author/Lee%2C+Chengkuo/loi/aplhttp://dx.doi.org/10.1063/1.4980115http://aip.scitation.org/toc/apl/110/16http://aip.scitation.org/publisher//doi/abs/10.1063/1.4980087/doi/abs/10.1063/1.4978672/doi/abs/10.1063/1.4980018/doi/abs/10.1063/1.4982164/doi/abs/10.1063/1.4976804/doi/abs/10.1063/1.4979543
Active MEMS metamaterials for THz bandwidth control
Kailing Shih,1,2,3,4 Prakash Pitchappa,1,2,4 Manukumara Manjappa,5 Chong Pei Ho,1,2,3,4
Ranjan Singh,5 Bin Yang,3 Navab Singh,6 and Chengkuo Lee1,2,4,7,a)1Department of Electrical and Computer Engineering, National University of Singapore, Singapore 1175762NUS Suzhou Research Institute (NUSRI), Suzhou, Industrial Park, Suzhou 215123,People’s Republic of China3Department of Micro/Nano Electronics, Shanghai Jiao Tong University, Shanghai 200240,People’s Republic of China4Center for Intelligent Sensors and MEMS (CISM), National University of Singapore, Singapore 1175765Division of Physics and Applied Physics, Centre for Disruptive Photonic Technologies, School of Physicaland New Mathematical Sciences, Nanyang Technological University, Singapore 6373716Institute of Microelectronics (IME), 11 Science Park Road, Singapore 1176857Graduate School for Integrative Science and Engineering, National University of Singapore,Singapore 117456
(Received 12 January 2017; accepted 2 April 2017; published online 19 April 2017)
We experimentally demonstrate a microelectromechanical system (MEMS) based metamaterial with
actively tunable resonance bandwidth characteristics, operating in the terahertz (THz) spectral region.
The broadband resonance characteristic feature of the MEMS metamaterial is achieved by integrating
sixteen microcantilever resonators of identical lengths but with continuously varying release lengths,
to form a supercell. The MEMS metamaterial showed broadband resonance characteristics with a full
width half maximum (FWHM) value of 175 GHz for resonators with a metal thickness of 900 nm and
was further improved to 225 GHz by reducing the metal thickness to 500 nm. The FWHM resonance
bandwidth of the MEMS metamaterial was actively switched to 90 GHz by electrostatically
controlling the out-of-plane release height of the constituent microcantilever resonators. Furthermore,
the electrically controlled resonance bandwidth allows for the active phase engineering with relatively
constant intensity at a given frequency based on the reconfiguration state of the MEMS metamaterial.
This enables a pathway for the realization of actively controlled transmission or reflection based on
dynamically programmable THz metamaterials. Published by AIP Publishing.[http://dx.doi.org/10.1063/1.4980115]
Electromagnetic (EM) metamaterials are arrays of sub-
wavelength structures, which have been shown to have inter-
esting EM properties, such as artificial magnetism,1 negative
refractive index,2 wavelength selective absorption,3 narrow
band emission,8 classical analogue of electromagnetically
induced transparency (EIT),4–7 Fano resonance,9 and many
more. The disruptive feature of metamaterials comes from the
possibility of achieving varied EM properties by engineering
the unit cell geometry. However, most of these reported meta-
materials are limited to provide a narrow band response,
which greatly hinder their usage from applications desiring
broadband characteristics such as broadband cloaking, imag-
ing, and spectrometry. Few earlier reports on broadband reso-
nance metamaterials realized this by integrating multiple
overlapping narrow band resonators into one supercell.10–17
The constituent resonators in the supercell are either placed in
a planar configuration with a lateral spatial variation15 or
stacked on top of one another.16 Alternatively, multiple modes
of the complex resonator design such as nested split ring reso-
nators are also reported to achieve broadband resonance.14
The planar broadband metamaterial is easier to fabricate but
provides lower resonance strength due to the reduced number
of resonators in a given area. On the other hand, the stacked
resonator metamaterial is more difficult to fabricate but pro-
vides stronger resonance strength. However, either of these
broadband resonance metamaterials reported so far are usually
passive or can achieve active amplitude modulation of broad-
band resonance.18 The active tunability of the resonance band-
width in terahertz (THz) metamaterials is yet to be reported
and will be a critical component in the realization of dynamic
bandwidth allocation for bandwidth control in next generation
sub-THz wireless communication systems.
The active control of the metamaterial response is
of immense importance for the realization of various metade-
vices, which will enable numerous practical applications.19
Conventionally, the active control of the metamaterial
response is achieved by integrating active materials into
the resonator geometry or as a surrounding medium, whose
properties can be controlled through external stimulus
such as optical,20,21 electrical,22 thermal,23,24 and magnetic
fields.25 However, the use of active materials over the wide
spectral range is limited due to their frequency dependent
material properties, complexity in the fabrication process,
and need for bulky systems to provide the external control.
Alternatively, structural reconfiguration of the resonator
geometry by integrating microelectromechanical system
(MEMS) based microactuators has been reported as an effi-
cient means of achieving tunable metamaterials especially
in THz18,26–32 and infrared33–35 spectral regions. In recent
years, MEMS metamaterials have been reported for the
active control of numerous EM properties such as magnetic
response,27,36 electrical response,32,37 multiband switch-
ing,38,39 EIT analogue,40–42 phase engineering,43 and
a)Author to whom correspondence should be addressed. Electronic mail:
elelc@nus.edu.sg
0003-6951/2017/110(16)/161108/5/$30.00 Published by AIP Publishing.110, 161108-1
APPLIED PHYSICS LETTERS 110, 161108 (2017)
http://dx.doi.org/10.1063/1.4980115http://dx.doi.org/10.1063/1.4980115mailto:elelc@nus.edu.sghttp://crossmark.crossref.org/dialog/?doi=10.1063/1.4980115&domain=pdf&date_stamp=2017-04-19
chirality.44 Versatility of the microactuator design, ease of
process integration, and electrical control makes it an ideal
candidate for realizing high performance and miniaturized
metadevice systems.
In this paper, we experimentally demonstrate a MEMS
metamaterial with electrostatically tunable bandwidth char-
acteristics operating in the THz spectral region. The MEMS
metamaterial supercell consists of 4� 4 arrays of cantileverdipolar resonators of identical total length (60 lm) and width(5 lm) but with gradually varying release lengths. Thisallows for the realization of the broadband response in the
initial OFF state, when no voltage is applied. When voltage
is applied between the released cantilever and silicon sub-
strate, the out-of-plane gap is closed, and all sixteen cantile-
vers resonate at the same resonance frequency, leading to a
narrow band response in the ON state. The proposed MEMS
metamaterial with bandwidth tunability can be adopted in
realization of next generation THz wireless communication
channels, THz imaging, and spectroscopic applications.
The proposed broadband MEMS metamaterial supercell
consists of an array of sixteen cantilever resonators with grad-
ually varying release lengths (Fig. 1(a)). The cantilever has
fixed part of length lf and released part of length lr, which are
made of bimaterial layers—tAl the thickness of aluminum
(Al) on top of td the thickness of aluminum oxide (Al2O3)
(Fig. 1(b)). The cantilever with no release part is termed as
C1, and on the other hand, the cantilever that is completely
released is termed as C16. All other cantilevers C2–C15 have
gradually varying released lengths with a step value of 4 lm.The optical microscopy image of sixteen cantilever resonators
is shown in Fig. 1(c). The released part of the cantilever is
bent in the out-of-plane direction with the tip displacement
(g), which is due to the residual stress in the bimaterial
layers.30,45 When THz waves with an electric field aligned
along the cantilever length is incident on the metamaterial
formed with a single cantilever as a unit cell, narrow dipolar
resonance is achieved.37 The resonance frequency of the can-
tilever metamaterial is given by fr � (Leff �Ceff)�1/2, whereLeff and Ceff are the effective inductance and capacitance of
the cantilever resonator, respectively. Ceff is a series
combination of capacitance from the Al2O3 layer (Cd) and
out-of-plane gap (Cg), i.e., Ceff a (Cd �Cg)/(CdþCg). The out-of-plane gap capacitance is given by Cg a (lr �wc)/g, and sobased on the values of lr and g, the cantilever dipolar reso-
nance can be engineered accordingly. Even with cantilever
resonators of equal total length, the dipolar resonances can be
altered by exploiting the third dimension of metamaterial
design to achieve overlapping resonance bands that can be
integrated to enable planar broadband resonance
metamaterials.
The effects of lr and g on the fr of the cantilever resonators
were independently studied by finite difference time domain
(FDTD) modeling. The out-of-plane gap is quantified in the
simulations by using a parameter termed as release angle,
h¼ sin�1(g/lr). First, metamaterials with a single cantileverresonator at fixed h and varying lr were studied. The simulatedTHz transmission spectra of the cantilever metamaterials at
h¼ 1� and varying lr are shown in Fig. 2(a). The dipolar reso-nance frequency of C1 (fr_C1) was observed at 0.58 THz. As lrincreases, the corresponding fr redshifts gradually and the nar-
rowest band fr is achieved for C16 (fr_C16 ¼ 0.705 THz).Hence, by integrating all sixteen cantilevers into a supercell,
broadband resonance can be achieved by the superposition of
sixteen dipolar resonances (dashed curve in Fig. 2(a)). The
maximum resonance frequency shift between the cantilevers
forming a supercell is calculated as Df¼ fr_C16� fr_C1. It isalso important to note that the resonance strength decreases
FIG. 1. (a) Schematics of a bandwidth tunable MEMS metamaterial super-
cell formed by a 4� 4 array of microcantilever resonators with varying fixedand released lengths. (b) Schematic of a microcantilever resonator with geo-
metrical definitions. (c) Optical microscopy image of the fabricated micro-
cantilever resonators (C1–C16) with varying fixed and released lengths.
FIG. 2. (a) Simulated microcantilever metamaterial THz transmission spec-
tra with varying release lengths, lr, at a release angle of h¼ 1�. The blackdashed line shows the Lorentz fitted broadband resonance spectrum of the
supercell formed by combining all sixteen microcantilever unit cells with
varying lr into a single supercell, and the inset shows the geometrical defini-
tions of lr, d, and h. (b) Simulated microcantilever resonator based metama-terial THz transmission spectra with varying lr and h. The inset shows thecalculated maximum resonance frequency shift between microcantilever
resonators with minimum lr� 0 lm (C1) and maximum lr ¼ 60 lm (C16) asDf¼ fr_C16� fr_C1 at varying h.
161108-2 Shih et al. Appl. Phys. Lett. 110, 161108 (2017)
with increasing lr due to reduced electric field confinement in
the out-of-plane gap. Furthermore, simulations were carried
out for varying h from 1� to 6� (Fig. 2(b)). Since hC1 is always0�, the fr_C1 remains unchanged. However, for the other canti-levers, the increment of fr at any given lr decreases with
increasing h. This is because of the large out-of-plane gapbetween the cantilever and substrate, leading to weaker reso-
nances at the higher frequency range in the unit cell.37,39 The
Df was calculated for h ¼ 1� to 6� and is shown in the inset ofFig. 2(b). The Df for h¼ 6� is calculated to be 0.275 THz,which is 120% higher than that for the h¼ 1� case with0.175 THz. Furthermore, the saturating behavior of Df athigher values of h is clearly observed. This defines the maxi-mum bandwidth that is achievable in the MEMS metamaterial.
Hence, from the simulation results, it is obvious that the pro-
posed approach of selectively releasing a part of the cantilever
provides an added dimensionality for the realization of broad-
band resonance metamaterials.
The proposed broadband MEMS metamaterial was fabri-
cated using a CMOS compatible process (tAl¼ 900 nm;td¼ 50 nm)32 with 100 lm pitch between cantilevers in anarea of 1 cm2 (supplementary material). Sixteen cantilevers
are positioned in a manner to have certain differences in the
release length with their neighboring resonators, thus prevent-
ing undesired sequential effects as a super cell. The eight can-
tilevers with shorter lr (C1–C8) were electrically connected to
each other but isolated from the metal lines connecting the
other eight cantilevers with larger lr (C9–C16). This electrical
isolation allows for the independent actuation of two sets of
cantilevers to enable advanced control of the resonance band-
width. When voltage is applied between the Al layer of canti-
levers and the substrate, the electrostatic force pulls the
released cantilevers towards the substrate. After a particular
value of input voltage called pull-in voltage, the cantilevers
will come in contact with the substrate.46,47 This allows for
the active control of the gap capacitance Cg of the MEMS
metamaterial, and hence its resonance frequency can be con-
trolled with greater precision. The voltage applied to shorter
and longer lr cantilevers is termed as VR and VL, respectively.
The THz transmission response of the fabricated reconfigura-
ble broadband MEMS metamaterial was measured using a
THz-time domain spectrometry (THz-TDS) system with THz
waves incident normally on the sample and electric field
polarized along the cantilever length. The THz transmission
amplitude and phase of the MEMS metamaterial at various
reconfiguration states are shown in Figs. 3(a) and 3(b), respec-
tively. When VR¼VL ¼ 0 V, the cantilevers are suspendedwith varying lr and g, and hence broadband resonance is
observed. This experimentally verifies the proposed concept
of realizing broadband resonance in the THz metamaterial by
MEMS technology. As VR is selectively increased, the shorter
lr cantilevers move towards the substrate and blueshifts the
corresponding resonance frequencies. The resonance strength
increases in the lower frequency range due to the increased
number of resonators (C2–C8) with fr close to fr_C1. This
increased asymmetry in the cantilever resonance frequency
variation causes a slight increase in the overall resonance
bandwidth of the MEMS metamaterial. More importantly,
when VL is increased, strong reduction in the resonance band-
width can be observed. This is due to the significantly larger
change in the out-of-plane gap capacitance of C9–C16. When
VR¼VL¼ 40 V, all the sixteen cantilevers are identical withno gap below. This leads to a single narrow band resonance at
0.58 THz. Due to the increased number of resonators, the nar-
row band resonance strength is also significantly higher. The
full-width half maximum bandwidth (B) of the measured THz
spectra is determined by fitting the curve with the Lorentzian
function and deducing the B value. The B of the MEMS meta-
material, when VL¼VR¼ 0 V, was measured to be0.175 THz and was tuned to a narrow B equal to 0.09 THz
(Fig. 3(c)). The transmission phase (u) was also measured atvarious reconfiguration states (Fig. 3(b)). When the resonance
bandwidth was larger, the phase gradient (du/df) was lower,and as the bandwidth reduces, the phase gradient gets sharper
(Fig. 3(c)). Hence, the resonance bandwidth tunability indi-
rectly enables the control of phase gradient. More interest-
ingly, the change in the phase gradient can be engineered to
achieve specific phase values with relatively constant
FIG. 3. Measured (a) THz transmission
and (b) phase of the bandwidth tunable
MEMS metamaterial (BTMM) with a
metal thickness of 900 nm in various
reconfiguration states. (c) Measured
FWHM resonance bandwidth, B
(black-solid curve), and phase change
gradient, du/df (red-dash curve) in vari-ous reconfiguration states of the
BTMM. (d) Measured THz transmis-
sion (blue-dash curve) and phase
(green-solid) at 0.65 THz in various
reconfiguration states of the BTMM.
161108-3 Shih et al. Appl. Phys. Lett. 110, 161108 (2017)
ftp://ftp.aip.org/epaps/appl_phys_lett/E-APPLAB-110-030716
amplitude and can potentially be used as dynamically pro-
grammable meta-bits in coding metamaterials.48 The mea-
sured phase change and normalized transmission at 0.65 THz
in various reconfiguration states are shown in Fig. 3(d).
Although the phase change observed here is very small for
practical applications, the proposed design approach can be
adopted in a metal-insulator-metal configuration to achieve
larger and more accurate phase control.43
In order to experimentally study the h variation, twoother samples were fabricated with lower tAl of 700 nm and
500 nm. The three fabricated samples with tAl¼ 900 nm,700 nm, and 500 nm are termed as D900, D700, and D500,
respectively. The tip displacement (g) of all the cantilevers
for the three fabricated samples were measured using a
reflection digital holographic microscope. The measured g
strongly depends on the value of tAl as shown in Fig. 4(a).37
The g increases with increasing lr and reducing tAl. Unlike
the consideration made in the simulations, the values of both
lr and h vary at the same time in real devices. The increase ing of the cantilever with decreasing tAl is primarily caused
due to the reduction in spring constant of the cantilevers.
Hence, in experiments, the influence of h is studied by vary-ing tAl.
37 All three MEMS metamaterials were characterized
for their THz transmission response in various reconfigura-
tion states. The B value of the transmission spectra are
shown in Fig. 4(b). For D700, when VR¼VL¼ 0 V, the Bvalue is measured to be 0.225 THz, which is approximately
75% higher than that of D900 (B¼ 0.175 THz). This con-firms that the B of the MEMS metamaterial increases with
increasing h. Compared to the D900 sample, a similar trendof bandwidth tuning is observed in the case of D700. When
VL¼VR¼ 40 V, the resonance bandwidth of D700 was mea-sured to be 0.09 THz, which matches with that of D900 in
the same reconfiguration state. In this state, all the cantilevers
are identical since they are all in physical contact with the
substrate. The only difference between D700 and D900 is tAl,
which has negligible influence on their resonance frequencies
and corresponding bandwidths. More interestingly, in the case
of D500 when VL¼VR¼ 0 V, the resonance bandwidth ismeasured to be 0.225 THz, which is identical to that of D700.
Although, the g of all the cantilevers for D500 is higher than
that of D700, the resonance strength of the cantilevers espe-
cially with larger lr becomes extremely weak and hence does
not contribute to the resonance broadening effect.39 This
experimentally shows the saturating behavior of the increasing
resonance bandwidth with larger h values as predicted in thesimulations. However, the trend of resonance bandwidth tun-
ing with the same order of reconfiguration states is identical to
all three samples. For D500 when VL¼VR¼ 40 V, the B wasapproximately 123 GHz, which is higher than that of both
D700 and D900. This could be caused due to the higher volt-
age required for completely closing the out-of-plane gap
for cantilevers with larger g. By increasing the applied volt-
age, the bandwidth of the D500 can also be narrowed to
90 GHz. Although in simulation, the effects of lr and h arestudied independently, in the real case, their variations mutu-
ally depend on each other. This leads to the slight difference
between the simulated and measured THz transmission spec-
tra. Furthermore, the saturating resonance broadening can be
overcome by varying the total length of cantilevers, which is
not explored here. The proposed approach of the achieving
broadband THz response using the MEMS metamaterial can
also be adopted for the realization of spectral broadband
switching, polarization-insensitive tunable bandwidth systems,
and broadband tunable absorbers that would enable a wide
range of potential THz applications.
In summary, a broadband MEMS metamaterial with an
electrically tunable bandwidth is experimentally demonstrated
by integrating cantilever resonators of varying release lengths
into one supercell. The measured FWHM of 0.175 THz was
achieved, which was reduced to 0.09 THz. The broadband res-
onance width was further increased by 120% by reducing the
Al thickness from 900 nm to 700 nm. However, with a further
reduction in the Al thickness to 500 nm, no obvious resonance
broadening was observed. These MEMS metamaterials are
fabricated using the CMOS compatible process and can be
easily integrated to application specific to ICs to enable elec-
trical control signaling. A similar approach can be applied to
metamaterial resonator designs and hence can enable the
whole class of interesting THz properties with a broadband
spectral response. This makes the proposed MEMS metama-
terial to be an ideal candidate for realization of numerous
miniaturized broadband THz systems and a critical compo-
nent for bandwidth management in next generation sub-THz
wireless communication systems.
See supplementary material for a detailed description of
the simulation method, the sample preparation, and the
device performance.
FIG. 4. (a) Measured tip displacement of the bandwidth tunable MEMS
metamaterials—D900 (black-square), D700 (red-circle), and D500 (blue-
triangle), respectively. (b) Measured FWHM resonance bandwidth of D900
(black-square), D700 (red-circle), and D500 (blue-triangle) at various recon-
figuration states.
161108-4 Shih et al. Appl. Phys. Lett. 110, 161108 (2017)
ftp://ftp.aip.org/epaps/appl_phys_lett/E-APPLAB-110-030716
The authors acknowledge the financial support from the
research grant (No. NRF-CRP001-019) at the National
University of Singapore and partially supported by the
National Natural Science Foundation of China under Grant
No. 61474078 at the NUS (Suzhou) Research Institute,
Suzhou, China. M.M. and R.S. acknowledge the research
funding support from NTU startup Grant No. M4081282,
Singapore, MOE Grant Nos. M4011362, MOE2011-T3-1-
005, and MOE2015-T2-2-103.
1J. B. Pendry, A. J. Holden, D. Robbins, and W. Stewart, IEEE Trans
Microwave Theory Tech. 47, 2075 (1999).2R. A. Shelby, D. R. Smith, and S. Schultz, Science 292, 77 (2001).3N. Landy, S. Sajuyigbe, J. Mock, D. Smith, and W. Padilla, Phys. Rev.
Lett. 100, 207402 (2008).4R. Singh, C. Rockstuhl, F. Lederer, and W. Zhang, Phys. Rev. B 79,085111 (2009).
5P. Tassin, L. Zhang, T. Koschny, E. Economou, and C. M. Soukoulis,
Phys. Rev. Lett. 102, 053901 (2009).6M. Manjappa, S.-Y. Chiam, L. Cong, A. A. Bettiol, W. Zhang, and R.
Singh, Appl. Phys. Lett. 106, 181101 (2015).7M. Manjappa, Y. K. Srivastava, and R. Singh, Phys. Rev. B 94, 161103(2016).
8X. Liu, T. Tyler, T. Starr, A. F. Starr, N. M. Jokerst, and W. J. Padilla,
Phys. Rev. Lett. 107, 045901 (2011).9R. Singh, I. A. Al-Naib, Y. Yang, D. R. Chowdhury, W. Cao, C. Rockstuhl,
T. Ozaki, R. Morandotti, and W. Zhang, Appl. Phys. Lett. 99, 201107 (2011).10J. Han, J. Gu, X. Lu, M. He, Q. Xing, and W. Zhang, Opt. Express 17,
16527 (2009).11Y. Q. Ye, Y. Jin, and S. He, JOSA B 27, 498 (2010).12S. Gu, J. Barrett, T. Hand, B.-I. Popa, and S. Cummer, J. Appl. Phys. 108,
064913 (2010).13N. Han, Z. Chen, C. Lim, B. Ng, and M. Hong, Opt. Express 19, 6990 (2011).14D. R. Chowdhury, R. Singh, M. Reiten, H.-T. Chen, A. J. Taylor, J. F.
O’Hara, and A. K. Azad, Opt. Express 19, 15817 (2011).15S. Gu, B. Su, and X. Zhao, J. Appl. Phys. 114, 163702 (2013).16F. Ding, Y. Jin, B. Li, H. Cheng, L. Mo, and S. He, Laser Photonics Rev.
8, 946 (2014).17L.-H. Gao, Q. Cheng, J. Yang, S.-J. Ma, J. Zhao, S. Liu, H.-B. Chen, Q.
He, W.-X. Jiang, and H.-F. Ma, Light: Sci. Appl. 4, e324 (2015).18M. Unlu and M. Jarrahi, Opt. Express 22, 32245 (2014).19N. I. Zheludev and Y. S. Kivshar, Nat. Mater. 11, 917 (2012).20H.-T. Chen, J. F. O’Hara, A. K. Azad, A. J. Taylor, R. D. Averitt, D. B.
Shrekenhamer, and W. J. Padilla, Nat. Photonics 2, 295 (2008).21M. Manjappa, Y. K. Srivastava, L. Cong, I. Al-Naib, and R. Singh, Adv.
Mater. 29, 1603355 (2017).22H.-T. Chen, W. J. Padilla, J. M. Zide, A. C. Gossard, A. J. Taylor, and
R. D. Averitt, Nature 444, 597 (2006).
23Q.-Y. Wen, H.-W. Zhang, Q.-H. Yang, Y.-S. Xie, K. Chen, and Y.-L. Liu,
Appl. Phys. Lett. 97, 021111 (2010).24R. Singh, A. K. Azad, Q. Jia, A. J. Taylor, and H.-T. Chen, Opt. Lett. 36,
1230 (2011).25L. Kang, Q. Zhao, H. Zhao, and J. Zhou, Opt. Express 16, 8825
(2008).26H. Tao, A. Strikwerda, K. Fan, W. Padilla, X. Zhang, and R. Averitt, Phys.
Rev. Lett. 103, 147401 (2009).27W. M. Zhu, A. Q. Liu, X. M. Zhang, D. P. Tsai, T. Bourouina, J. H. Teng,
X. H. Zhang, H. C. Guo, H. Tanoto, and T. Mei, Adv. Mater. 23, 1792(2011).
28W. Zhu, A. Liu, W. Zhang, J. Tao, T. Bourouina, J. Teng, X. Zhang, Q.
Wu, H. Tanoto, and H. Guo, Appl. Phys. Lett. 99, 221102 (2011).29W. Zhu, A. Liu, T. Bourouina, D. Tsai, J. Teng, X. Zhang, G. Lo, D.
Kwong, and N. Zheludev, Nat. Commun. 3, 1274 (2012).30Z. Han, K. Kohno, H. Fujita, K. Hirakawa, and H. Toshiyoshi, Opt.
Express 22, 21326 (2014).31C. P. Ho, P. Pitchappa, Y.-S. Lin, C.-Y. Huang, P. Kropelnicki, and C.
Lee, Appl. Phys. Lett. 104, 161104 (2014).32F. Ma, Y.-S. Lin, X. Zhang, and C. Lee, Light: Sci. Appl. 3, e171 (2014).33X. Liu and W. J. Padilla, Adv. Opt. Mater. 1, 559 (2013).34P. Pitchappa, C. P. Ho, P. Kropelnicki, N. Singh, D.-L. Kwong, and C.
Lee, Appl. Phys. Lett. 104, 201114 (2014).35N. I. Zheludev and E. Plum, Nat. Nanotechnol. 11, 16 (2016).36Z. Han, K. Kohno, H. Fujita, K. Hirakawa, and H. Toshiyoshi, IEEE J.
Sel. Top. Quantum Electron. 21, 2700809 (2015).37P. Pitchappa, C. P. Ho, L. Dhakar, Y. Qian, N. Singh, and C. Lee,
J. Microelectromech. Syst. 24, 525 (2015).38P. Pitchappa, C. P. Ho, L. Dhakar, and C. Lee, Optica 2, 571
(2015).39P. Pitchappa, C. P. Ho, Y. Qian, L. Dhakar, N. Singh, and C. Lee, Sci.
Rep. 5, 11678 (2015).40P. Pitchappa, M. Manjappa, C. P. Ho, R. Singh, N. Singh, and C. Lee,
Adv. Opt. Mater. 4, 541 (2016).41P. Pitchappa, M. Manjappa, C. P. Ho, Y. Qian, R. Singh, N. Singh, and C.
Lee, Appl. Phys. Lett. 108, 111102 (2016).42P. Pitchappa, M. Manjappa, C. P. Ho, R. Singh, N. Singh, and C. Lee,
Appl. Phys. Lett. 109, 211103 (2016).43L. Cong, P. Pitchappa, Y. Wu, L. Ke, C. Lee, N. Singh, H. Yang, and R.
Singh, Adv. Opt. Mater. 5, 1600716 (2017).44T. Kan, A. Isozaki, N. Kanda, N. Nemoto, K. Konishi, H. Takahashi, M.
Kuwata-Gonokami, K. Matsumoto, and I. Shimoyama, Nat. Commun. 6,8422 (2015).
45Y.-S. Lin, Y. Qian, F. Ma, Z. Liu, P. Kropelnicki, and C. Lee, Appl. Phys.
Lett. 102, 111908 (2013).46Y. Qian, L. Lou, M. J. Tsai, and C. Lee, Appl. Phys. Lett. 100, 113102
(2012).47P. Singh, C. G. Li, P. Pitchappa, and C. Lee, Electron Device Lett., IEEE
34, 987 (2013).48T. J. Cui, M. Q. Qi, X. Wan, J. Zhao, and Q. Cheng, Light: Sci. Appl. 3,
e218 (2014).
161108-5 Shih et al. Appl. Phys. Lett. 110, 161108 (2017)
http://dx.doi.org/10.1109/22.798002http://dx.doi.org/10.1109/22.798002http://dx.doi.org/10.1126/science.1058847http://dx.doi.org/10.1103/PhysRevLett.100.207402http://dx.doi.org/10.1103/PhysRevLett.100.207402http://dx.doi.org/10.1103/PhysRevB.79.085111http://dx.doi.org/10.1103/PhysRevLett.102.053901http://dx.doi.org/10.1063/1.4919531http://dx.doi.org/10.1103/PhysRevB.94.161103http://dx.doi.org/10.1103/PhysRevLett.107.045901http://dx.doi.org/10.1063/1.3659494http://dx.doi.org/10.1364/OE.17.016527http://dx.doi.org/10.1364/JOSAB.27.000498http://dx.doi.org/10.1063/1.3485808http://dx.doi.org/10.1364/OE.19.006990http://dx.doi.org/10.1364/OE.19.015817http://dx.doi.org/10.1063/1.4826911http://dx.doi.org/10.1002/lpor.201400157http://dx.doi.org/10.1038/lsa.2015.97http://dx.doi.org/10.1364/OE.22.032245http://dx.doi.org/10.1038/nmat3431http://dx.doi.org/10.1038/nphoton.2008.52http://dx.doi.org/10.1002/adom.201600417http://dx.doi.org/10.1002/adom.201600417http://dx.doi.org/10.1038/nature05343http://dx.doi.org/10.1063/1.3463466 http://dx.doi.org/10.1364/OL.36.001230http://dx.doi.org/10.1364/OE.16.008825http://dx.doi.org/10.1103/PhysRevLett.103.147401http://dx.doi.org/10.1103/PhysRevLett.103.147401http://dx.doi.org/10.1002/adma.201004341http://dx.doi.org/10.1063/1.3664131http://dx.doi.org/10.1038/ncomms2285http://dx.doi.org/10.1364/OE.22.021326http://dx.doi.org/10.1364/OE.22.021326http://dx.doi.org/10.1063/1.4871999http://dx.doi.org/10.1126/science.1133628http://dx.doi.org/10.1002/adom.201300163http://dx.doi.org/10.1063/1.4879284http://dx.doi.org/10.1038/nnano.2015.302http://dx.doi.org/10.1109/JSTQE.2014.2378591http://dx.doi.org/10.1109/JSTQE.2014.2378591http://dx.doi.org/10.1109/JMEMS.2015.2421307http://dx.doi.org/10.1364/OPTICA.2.000571http://dx.doi.org/10.1038/srep11678http://dx.doi.org/10.1038/srep11678http://dx.doi.org/10.1002/adom.201500676http://dx.doi.org/10.1063/1.4943974http://dx.doi.org/10.1063/1.4969061http://dx.doi.org/10.1002/adom.201600716http://dx.doi.org/10.1021/ja9938581http://dx.doi.org/10.1063/1.4798244http://dx.doi.org/10.1063/1.4798244http://dx.doi.org/10.1063/1.3693382http://dx.doi.org/10.1109/LED.2013.2262918http://dx.doi.org/10.1070/PU1968v010n04ABEH003699
ln1f1f2f3f4c1c2c3c4c5c6c7c8c9c10c11c12c13c14c15c16c17c18c19c20c21c22c23c24c25c26c27c28c29c30c31c32c33c34c35c36c37c38c39c40c41c42c43c44c45c46c47c48
Recommended